Diagnostic Technologies to Assess Tissue Perfusion and Cardiorespiratory Performance

Introduction

Fundamental to the management of critically ill patients is the on-going assess- ment of both cardiorespiratory status and the adequacy of tissue perfusion. How- ever, the management of the critically ill is also context specifi c. What measures one makes in the operating room, where tight titration of support in the face of surgical trauma is the rule, to fi eld and Emergency Department settings where invasive monitoring is impractical and steady state conditions rarely present, limit the generalizability of statements about specifi c devices and their utility.

Furthermore, no monitoring device will improve outcome unless coupled to a treatment, which, itself, improves outcome. The next fi ve years should witness a closer synchrony between monitoring techniques and goals in terms of pa- tient-centered outcomes. Within the context, some truths have sustained clini- cal scrutiny and speak to consistent treatment logic that should progress more over the next fi ve years, although they will probably not develop into mature applications.

Tissue hypoperfusion, or circulatory shock, induces a profound sympathetic response that aims to restore central arterial pressure to sustain cerebral and coronary blood fl ow. The body does this at the expense fi rst of the skin, non-ac- tive muscles and renal ultrafi ltration, followed closely by splanchnic vasocon- striction [1]. Fundamentally, therefore, a blood pressure within the normal range and the presence of mentation in a patient do not insure cardiovascular suffi - ciency. Regrettably, the opposite is true. In previously healthy trauma patients, mentation and mean arterial pressures in excess of 90 mmHg often exist within the context of profound hypoperfusion and ischemic injury to the gut mucosa and renal medulla. Such states are often referred to as compensated shock, to describe this masquerade. Furthermore, delayed resuscitation even if it restores regional blood fl ow, may not prevent organ injury or rapidly restore organ func- tion. The increased morbidity and mortality from delayed or inadequate resus- citation efforts have been documented in the trauma literature [2].

Finally, monitoring and resuscitation are also context specifi c. Differing groups of patients refl ect differing needs and can be assessed well to differing degrees. One must apply differing fi lters to the review of the literature and the application of technologies in assessing tissue perfusion and monitoring tech- niques based on a variety of criteria. However, here we shall discuss several new

and Cardiorespiratory Performance

M. R. Pinsky

and exciting developments that should become commonplace within the next fi ve years. These advances can be broadly placed in the following categories:

1) continuous non-invasive or minimally invasive measures of tissue wellness;

2) assessments of cardiopulmonary reserve and responsiveness to therapies;

and, fi nally

3) protocolized care to minimize medical errors, practice variation and apply best practices across broad groups of patients independent of the training of the bedside healthcare provider, sophistication of the intensive care unit (ICU) and patient population differences.

Within the framework monitoring of tissue perfusion and tissue wellness will become increasingly integrated into treatment protocols, defi ning not only who is ill but also the end-points of therapy.

Assessment of Tissue Wellness

A reasonable therapeutic goal is to restore blood fl ow to tissues so as to reverse tissue hypoperfusion, progressive ischemic cellular injury and their associated organ dysfunction. Preventing sustained tissue hypoperfusion is usually associ- ated with a far better outcome than treatment strategies that allow tissue hypop- erfusion to persist to the point of inducing organ injury or if the resuscitation is inadequate so that tissue perfusion is not fully restored to a level that will result in no perfusion defi cit. However, tissue perfusion is a relative concept. Tissues need only as much blood fl ow as required to meet their metabolic demands. Al- though one can make the distinction between fl ow and oxygen delivery, under most conditions these two variables are coupled. Assessment of tissue wellness usually revolves around estimates of the adequacy of aerobic metabolism and the associated performance of the tissue being monitored. Measures of tissue or venous PCO

are often used to assess blood fl ow because CO

production is remarkably constant for all organs, except the kidney, as blood fl ow decreases up until the point that oxygen extraction cannot sustain oxidative phosphorylation [3]. Numerous probes have been developed and studied to assess oxygen suf- fi ciency, regional blood fl ow for Doppler or CO

fl ux, and mitochondrial energy state.

Oxygen Sufficiency

Of all the measures of regional perfusion, oxygen measures are the least accurate

because they rely on too many unproven assumptions. On a global level, one can

measure mixed venous oxygen saturation (SvO

). Assuming arterial oxygen sat-

uration (SaO

) is > 92% then SvO

will be a function of the oxygen carrying ca-

pacity, the mean weighted fl ow to all organs and metabolic demand. Over a short

time interval, hemoglobin and hemoglobin oxygen binding affi nity usually do

not change, and if metabolic demand is controlled, then one can use SvO

as a

measure of global oxygen delivery (DO

) suffi ciency. Under normal conditions

SvO

stays around 72%. However, if it becomes ≤ 70% then circulatory stress is occurring and, if SvO

becomes < 65%, some areas of tissue ischemia probably exist [4]. However, an SvO

> 72% does not insure that all tissues have adequate blood fl ow because shunt fl ow may artifi cially increase end-capillary PO

in hy- perperfused regions masking tissue ischemia. In fact, this is often the argument given for why organ dysfunction is seen in septic patients when their SvO

is >

72%, cardiac output elevated, but hyperlactatemia is present. Although other reasonable explanations for these fi ndings can be given, including mitochondria block, decreased lactate clearance and failure of lactate dehydrogenase to con- vert lactate to pyruvate [5], clear documentation of microcirculatory derange- ments in septic patients has also been described [6]. Decreases in SvO

have also been used to identify those ventilatory-dependent subjects who will fail to wean from mechanical ventilatory support, as weaning successes did not display a de- crease in SvO

during their spontaneous breathing trials [7].

Recently, interest in using central venous SO

(ScvO

) as an alternative to SvO

has been proposed, because ScvO

can be measured from a central venous catheter, is similar though not identical to SvO

and eliminates the need for pul- monary arterial catheterization to measure SvO

[8, 9]. A resuscitation protocol based on ScvO

as a surrogate of SvO

resulted in better outcome from septic shock in patients treated in an emergency department than similar patients treated without the aid of ScvO

or SvO

guided therapy [10]. Thus, measures of compensated shock, even if measured using degraded measures, such as ScvO

, may still improve outcome if coupled to an aggressive treatment protocol. Al- though this approach has been endorsed by the SCCM and ESICM Surviving Sepsis campaign for the treatment of patients with septic shock [11], the pub- lished protocol [10] did not treat acutely ill in-patients within an ICU, but sub- jects presenting to an emergency department whose entire protocolized treat- ment was done there. Furthermore, care once in the ICU was not controlled and this emergency department protocol used very large red blood cell transfusions independent of fl uid resuscitation, which could have independently altered out- come. Furthermore, using threshold values of ScvO

to drive these protocols is problematic, since the relation between ScvO

and SvO

may vary markedly in shock states and during therapy [12]. However, this study clearly documented that using measures of oxygen extraction stress as a marker of compensated shock identifi ed patients whose outcome improved if resuscitation persisted fur- ther to resolve this presumed defi ciency.

Although regional measures of tissue PO

are possible, including splanch-

nic oxygen consumption [13], in-dwelling oxygen-sensitive electrode catheters

placed into tissues [14] and non-invasive measures of tissue PO

[15], the clinical

utility of these devices has not been shown and their use has been limited to a

few research centers. When muscle PO

levels are measured during experimen-

tal hemorrhagic shock and resuscitation, the trends appear to refl ect ischemia

and its recovery. Thus, measures of oxygen suffi ciency exist and on a global level

have proven useful to drive therapy in one clinical trial. Still, the sensitivity of

global measures of oxygen suffi ciency is poor, whereas regional measures of

tissue PO

have not been used to drive treatment protocols. Finally, in subjects

with prolonged cardiovascular insuffi ciency, as may occur in patients with se-

vere sepsis and septic shock, the potential exists that mitochondrial dysfunction may develop owing to impaired mitochondrial protein synthesis needed to drive oxidative phosphorylation [16]. Under these conditions, otherwise adequate tis- sue PO

levels (i.e., > 10 mmHg) may still be associated with impaired energy metabolism. However, under these circumstances of metabolic block, both oxy- gen consumption and CO

production will be limited. To date, no studies have reported decreased CO

production as part of high output circulatory shock.

Regional Blood Flow

Regional blood fl ow is heterogeneous both among organs and within organs, and regional blood fl ow distribution may vary over time in response to changes in sympathetic tone and treatments [17]. However, under most circumstances, regional blood fl ow is proportional to regional metabolic demand. Since re- gional metabolic demand refl ects oxidative phosphorylation, CO

production is also proportional to regional blood fl ow. Thus, tissue PCO

remains remarkably constant across tissues as metabolic rate varies as long as blood fl ow can co-vary with metabolic rate. Since total PCO

is also dependent on tissue CO

stores, measures of a specifi c threshold value for tissue PCO

are less informative than are measures of the differences between tissue PCO

and arterial PCO

. This so- called regional PCO

gap can be used to measure regional blood fl ow in the gut, if measured by gastric tonometry [18], the mouth, if measured by sublingual PCO

[19] in muscle and subcutaneous tissues, if measured by indwelling PCO

elec- trodes [20], and in the bladder musoca using surface electrodes [21]. Gutierrez et al. used an elevated gastric PCO

gap equivalent to identify compensated shock in critically ill subjects following an initial resuscitation [22]. Others have shown that gastric tonometry can identify occult circulatory failure during weaning [23]. Subjects who fail to wean from mechanical ventilatory support also develop an elevated gastric PCO

gap, consistent with gastric vasoconstriction during the stress of weaning [7]. Interestingly, when PCO

measures are coupled with pH measures, one has the unique opportunity to access the emergency of regional tissue metabolic acidosis, as the point wherein tissue pH decreases more rapidly than can be explained by the associated increase in PCO

based on the Hender- son-Hasselbalch equation. Potentially coupled pH-PCO

measures will be used in the future to identify critical DO

states and the onset of metabolic acidosis.

Thus, measures of regional blood fl ow exist and may be more robust than meas- ures of oxygen suffi ciency. However, their primary limitation at the present is the diffi culty in acquiring reliable measures of PCO

and its change over time.

Regional gastric mucosal blood fl ow can also be measured by local laser Dop- pler techniques; although this technique is useful for research purposes, it has not been used to document results of resuscitation or drive resuscitation pro- tocols in humans. Presumably, this lack of study is due to the invasive nature of the technique and its instability over time. Finally echo-imaging techniques allow for the assessment of renal cortical blood fl ow and myocardial blood fl ow.

Similarly, trans-cranial Doppler measures are used to assess cerebral blood fl ow.

Along this same line, echo-derived estimates have been reported for numerous

regional vascular beds. However, although such echo-derived measures of re- gional blood fl ow can be used to describe large vessel blood fl ow, their ability to assess actual tissue blood fl ow or effective blood fl ow is limited.

Cellular Energy Status

Cellular energy production through metabolism of three-carbon units within the tricarboxylic acid cycle to produce CO

and reduced cytochromes needed to feed mitochondrial oxidative phosphorylation for the production of ATP and consumption of oxygen refl ects the basic means by which cells, organ and whole animals survive and adapt to the environment. Although anaerobic metabolism of glucose to pyruvate and then lactate can generate some ATP, this strategy is grossly ineffi cient and unable to sustain cellular metabolism at any but the most quiescent level. Accordingly, numerous techniques have been studied that aim to assess mitochondrial redox state. With increased hypoxic stress, oxidative phos- phorylation is impaired with increased intracellular reduced cytochrome a/a3 and NADH levels. Since adequate oxygen fl ux from capillaries to mitochondria represents the fi nal pathway of oxygen transport to the cells, measures of reduced cytochrome a/a3 and NADH levels represent a direct evaluation of the effec- tiveness of resuscitation to sustain cellular function [24]. Although circulatory shock states have been demonstrated to induce impaired intracellular oxidative phosphorylation, and resuscitation has been shown to improve these measures, they have not been used in clinical trials. The reasons for this are multiple, not the least of which is the reality that such measures are, by necessity, highly local- ized and marked regional differences in tissue oxygen suffi ciency often exists in both health and disease. Potentially, measuring these parameters non-invasively over a spectrum of tissues and sites within tissues may give a clearer picture of tissue wellness. The second major weakness of these approaches is the lack of a calibrated threshold below which injury occurs and above which tissue energy production is adequate. In essence, absolute levels of any oxidative phosphoryla- tion factor or co-factor are less important than the rate of ATP turnover. Levels of these agents are analogous to the number of wheels on a car, not the rate at which the car is moving. Furthermore, increased reduced cytochrome and co- factor levels, using this same analogy, refl ect the state of infl ation of those tyres, not their potential rate of turn given an adequate substrate.

Assessment of Cardiorespiratory Reserve and Responsiveness to Therapy Hypoxemia

Hypoxemia, defi ned as an SaO

< 92%, usually corresponds to an arterial PO

of approximately 65 mmHg. As arterial PO

decreases below this value SaO

rap- idly decreases owing to the shape of the hemoglobin oxygen dissociation curve.

Although chronic hypoxemia can be tolerated remarkably well, acute arterial de-

saturation usually causes end-organ dysfunction. Although one may accurately

measure SaO

during arterial blood gas analysis, this approach has been large- ly superseded by the indirect measure of SaO

using pulse oximetry to derive pulse oximetry saturation (SpO

). Pulse oximeters determine SO

by measuring the light absorption of arterial blood at two specifi c wavelengths, 660 nm (red) and 940 nm (infrared). While pulse oximetry is accurate in refl ecting one-point measurements of SaO

, it does not reliably predict changes in SaO

[25]. Moreo- ver, the accuracy of pulse oximeters deteriorates when SaO

falls to 80% or less.

Still, since the lungs and the body do not store suffi cient oxygen reserves to meet metabolic demands, changes in arterial oxygenation may be seen in changes in SpO

. Importantly, titration of positive-end expiratory pressure (PEEP) and in- spired oxygen fraction (FiO

) in patients with acute lung injury (ALI) is usually done to achieve a SpO

< 90% [26].

Subjects with acute exacerbations of chronic obstructive lung disease often have arterial desaturation. The etiology of this desaturation is often complex and includes ventilation/perfusion (V/Q) mismatching, increased metabolic demand and intrapulmonary shunting [27]. Importantly, exercise, by increas- ing both cardiac output and pulmonary arterial pressure tends to minimize V/Q mismatch while not affecting shunt. Since exercise also increases oxygen consumption, SvO

usually decreases as well. Since V/Q mismatch is amenable to small increases in FiO

and is not infl uenced greatly by low SvO

values, pa- tients with predominately V/Q mismatch will improve their SpO

with sitting up, whereas those patients with primarily fi xed intrapulmonary shunts will demon- strate a decrease in SpO

. The advantage of this simple test is that the treatments for V/Q mismatch and shunt are different, thus allowing a simple non-invasive measure to both make a diagnosis and defi ne therapy. Such provocative uses of non-invasive SpO

monitoring represent the new frontier of this technique, which has gone underutilized for too long.

Preload-responsiveness

Over the past few years it has become increasingly evident that static measure of left ventricular (LV) volumes or indirect estimates of LV preload, such as pulmo- nary artery occlusion pressure (PAOP) and right atrial pressure, do not refl ect actual LV end-diastolic volume or predict the subsequent change in LV stroke volume in response to volume loading [28]. However, other provocative moni- toring techniques have been documented to be very sensitive and robust param- eters of preload-responsiveness.

Volume challenge: The traditional method of assessing preload-responsiveness

in a hemodynamically unstable patient is to rapidly give a relatively small intra-

vascular bolus of volume and observe the subsequent hemodynamic response in

terms of blood pressure, pulse, cardiac output and related measures [11]. Volume

challenges, if given in too large a volume, carry the risk of inducing acute volume

overload with its associated right ventricular (RV) failure (acute cor pulmonale)

or LV overload (pulmonary edema). Importantly, several surrogate methods of

creating reversible or transient volume challenges include passive leg raising and using ventilatory maneuvers.

Passive leg raising: Passive leg raising transiently increases venous return [29].

In one study, subjects who demonstrate a sustained increase in mean cardiac output or arterial pressure 30 seconds after the legs were raised were found to be preload responsive [30]. The advantage of this approach is that it can be used in patients who are breathing spontaneously, on mechanical ventilation or partial ventilatory assist, and it can be used even in the setting of atrial fi brillation. One limitation of this technique is that the volume mobilized by leg raising is also dependent on total blood volume and could be small in severely hypovolemic patients.

Central venous pressure (CVP) and inferior vena caval (IVC) diameter changes during spontaneous ventilation: Since the primary determinant of preload-re- sponsiveness is RV performance, tests that specifi cally assess RV performance are inherently attractive. During spontaneous inspiration, the pressure gradient for venous return increases and intrathoracic pressure becomes more negative.

CVP will decrease in preload-responsive patients [31] because the right ventricle can accommodate the increased infl ow without over-distending. Similarly, IVC diameter measured in its intrahepatic position will collapse if CVP is < 10 mm Hg [32]. Unfortunately, the changes in CVP are quite small and may be within the error of measurement of most ICUs. Similarly, measures of IVC diameter re- quire an expert bedside echocardiographer to visualize the intra-hepatic IVC.

Changes in LV output during positive-pressure ventilation: During positive- pressure ventilation, changes in LV output refl ect combined RV-LV performance.

Positive-pressure inspiration transiently decreases venous return, causing LV fi lling to vary in a cyclic fashion as well. If the RV–LV system is preload-respon- sive, then these small cyclic changes must also alter LV stroke volume. Since the greater the increase in tidal volume, the greater the transient decrease in venous return and subsequently greater decrease in LV output [33], one can program a ventilator to deliver a series of increasing tidal volumes and assess the degree of stroke volume decrease as a measure of preload-responsiveness [34]. During fi xed tidal volume positive-pressure ventilation cyclic variations in systolic pres- sure [35], pulse pressure [36], LV stroke volume [37] or aortic fl ow [38] quantify preload-responsiveness. Several monitoring devices are available to measure these pressure and fl ow variations, although none have been studied prospec- tively as a guide to treatment. Importantly, since measures of systolic pressure, pulse pressure, aortic fl ow, and descending aortic fl ow and fl ow velocity all have different relations to the dynamic change in LV end-diastolic volume to stroke volume relation for the same subject, the threshold values for each parameter predicting preload-responsiveness must be different between parameters and may demonstrate different degrees of robustness in their clinical utility [39].

Dobutamine challenge: Assuming that some preload-reserve is present but lim-

ited by impaired cardiac contractility, then one may increase inotropic state as a

challenge and observe its effect of regional oxygenation. Using this test, Creteur et al. [40] demonstrated that occult splanchnic ischemia could be unmasked in approximately 30% of critically ill patients who were otherwise considered sta- ble.

Use of Protocolis Driven by Physiological End-Points to Standardize Resuscitation

The application of evidence-based guidelines to clinical practice has gained in- creased acceptance in recent years. Using a rigorous and open method of grad- ing the scientifi c evidence and then placing it within the framework on known physiology and local practice patterns can produce remarkable treatment algo- rithms that, though not perfect, even at their inception, are still better than the nearly random behavior of the present practice of medicine. There are obvious potential advantages of introducing decision-making that is rational and based on evidence. Clinical care can be simplifi ed by the use of treatment algorithms that may facilitate cooperation between different healthcare professionals. This approach may also reduce costs for healthcare purchasers [41]. Unfortunately, standardization of care by the use of guidelines and protocols may not necessar- ily be a success. There are no certainties in clinical care and individual patients may have needs that fall outside standard guidelines. The rigid application of protocols cannot replace clinical judgment. Furthermore, practice guidelines imposed from outside and without clear rational or perceived benefi t may not be accepted by the practicing physicians. However, by merging bedside monitoring technologies with clinical information technology that can assess tissue well- ness to run decision-support systems, real-time support and monitoring can be accomplished, and in a cost-effective fashion.

For example, fl uid optimization as an endpoint of resuscitation has been shown to reduce length of hospital stay and important complications in patients undergoing a variety of major surgical procedures [42]. This is usually achieved with an esophageal Doppler monitor to evaluate changes in fl ow in response to fl uid challenges [43–45]. Markers of tissue perfusion have also been used as end- points for resuscitation. These include SvO

, ScvO

, lactate, base excess and gut mucosal pH using gastric tonometry. The use of these endpoints is attractive, because ensuring optimal blood fl ow to the vital organs is the ultimate aim of hemodynamic resuscitation. Those patients who spontaneously achieve target hemodynamics fall into a low mortality group and those who are non-achievers despite treatment have a high mortality. These differences must be due to the underlying fi tness of patients [46]. Protocolized care plans should use best evi- dence, aim to reverse those processes known to increase morbidity and mortal- ity, and at the same time be simple enough to be applied across care centers.

Another such protocol that lends itself to this approach is referred to as Func-

tional Hemodynamic Monitoring, because it asks of the monitoring information

what treatments should be given, not what is the diagnosis [47]. One needs to

fi rst know that a patient is unstable with tissues at risk or existing in an ischemic

state. Once this has been identifi ed, then how to treat the patient can be greatly

simplifi ed. Fundamentally there are only three questions asked of the cardio- vascular system regarding response to resuscitation efforts during shock. First, will blood fl ow to the body increase (or decrease) if the patient’s intravascular volume is increased (or decreased), and if so, by how much? Second, is any de- creased in arterial pressure due to loss of vascular tone or merely due to inad- equate blood fl ow? And third, is the heart capable of maintaining an effective blood fl ow with an acceptable perfusion pressure without going into failure? If taken in this order, one can develop a logical treatment algorithm. If the patient is hemodynamically unstable and preload-responsive, then treatment must in- clude immediate volume expansion as part of its overall plan. However, if the patient is also hypotensive and has reduced vasomotor tone, then even if car- diac output were to increase with volume expansion, arterial pressure may not increase in parallel. Thus, vital organ blood fl ow may remain compromised de- spite increasing cardiac output. Since the goal of resuscitation is to restore tissue perfusion, knowing that perfusion pressure will decrease despite an increasing cardiac output, defi nes that the physician should also start a vasopressor agent in tandem with the initial fl uid resuscitation so that both pressure and fl ow in- crease. If the patient is not preload-responsive but has reduced vasomotor tone, then a vasopressor alone is indicated, since fl uid resuscitation will not improve organ perfusion and may precipitate RV or LV overload. In the patient with cir- culatory shock who is neither preload-responsive or displaying reduced vasomo- tor tone, the problem is the heart and both diagnostic and therapeutic actions must be taken to address these specifi c problems (e.g., echocardiography, dob- utamine). The exact cause of cardiac compromise is not addressed by this ap- proach. For example massive pulmonary embolism (acute cor pulmonale), acute hemorrhagic tamponade and massive myocardial infarction would all fall into this category but would have markedly different treatments. However, since the diagnostic development can rapidly default to this group within minutes of ini- tiating the algorithm, appropriate diagnostic approaches could then be used in a highly focused fashion. Although this simplifi ed algorithm has not been used in clinical trials, it has the advantage of being easy to defend on physiological grounds, as well as being simple and easy to apply at the bedside across patient subgroups.

Acknowledgement

The work was supported in part by NIH grants HL67181–02A1 and HL07820–

06

References

1. Schlichtig R, Kramer D, Pinsky MR (1991) Flow redistribution during progressive hemor- rhage is a determinant of critical O

delivery. J Appl Physiol 70:169–178

2. Ivatury RR, Sugerman H (2000) In quest of optimal resuscitation: tissue specifi c, on to the

microcirculation. Crit Care Med 28:3102–3103

3. Schlichtig R, Kramer DJ, Boston JR, Pinsky MR (1991) Renal 0

consumption is supply- dependent at all levels of renal 0

delivery during progressive hemorrhage. J Appl Physiol 70:1957–1962

4. Robin ED (1980) Of men and mitochondria: coping with hypoxic dysoxia. The 1980 J.

Burns Amberson Lecture. Am Rev Respir Dis 122:517–531

5. Pinsky MR (2004) Goals of resuscitation in circulatory shock. Contrib Nephrol 144:31–

43

6. Ince C, Sinaasappel M (1999) Microcirculatory oxygenation and shunting in sepsis and shock. Crit Care Med 27:1369–1377

7. Jabran A, Mathru M, Dries D, Tobin MJ (1998) Continuous recordings of mixed venous oxygen saturation during weaning from mechanical ventilation and the ramifi cations thereof. Am J Respir Crit Care Med 158:1763–1769

8. Scheinman MM, Brown MA, Rapaport E (1969) Critical assessment of use of central ve- nous oxygen saturation as a mirror of mixed venous oxygen in severely ill cardiac pa- tients. Circulation 40:165–172

9. Rady MY, Rivers EP, Martin GB, Smithline H, Appelton T, Nowak RM (1992) Continuous central venous oximetry and shock index in the emergency department: use in the evalu- ation of clinical shock. Am J Emerg Med 10:538–543

10. Rivers E, Nguyen B, Havstad S, et al (2001) Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 345:1368–1377

11. Dellinger RP, Carlet JM, Masur H, et al (2004) Surviving Sepsis Campaign Guidelines for management of severe sepsis and septic shock. Crit Care Med 32:858–873

12. Reinhart K, Kuhn HJ, Hartog C, et al (2004) Continuous central venous and pulmonary artery oxygen saturation monitoring in the critically ill. Intensive Care Med 30:1572–

1578

13. Creteur J, De Backer D, Vincent JL (1999) Does gastric tonometry monitor splanchnic perfusion? Crit Care Med 27:2480–2484

14. Boekstegers P, Weidenhofer S, Kapsner T, Werdan K (1994) Skeletal muscle partial pres- sure of oxygen in patients with sepsis. Crit Care Med 22:640–650

15. McKinley BA, Marvin RG, Cocanour CS, Moore FA (2000) Tissue hemoglobin O

satura- tion during resuscitation of traumatic shock monitored using near infrared spectrom- etry. J Trauma 48:637–642

16. Singer M, De Santis V, Vitale D, Jeffcoate W (2004) Multiorgan failure is an adaptive, en- docrine-mediated, metabolic response to overwhelming systemic infl ammation. Lancet 364:545–548

17. Eckert R, Randall D, Burggren W, French K (2002) Circulation. In: Randall D, Burggren W, French K (eds) Eckert Animal Physiology. Mechanisms and Adaptations, 5

ed. WH Freeman & Co, New York pp 473–523

18. Tang W, Weil MH, Sun S, Noc M, Gazmuri RJ, Bisera J (1994) Gastric intramural PCO

as a monitor of perfusion failure during hemorrhagic and anaphylactic shock. J Appl Physiol 76:572–577

19. Nakagawa Y, Weil MH, Tang W, et al (1998) Sublingual capnometry for diagnosis and quantifi cation of circulatory shock. Am J Respir Crit Care Med 157:1838–1843

20. Sims C SP, Menconi M, Matsuda A, Pettit J, Puyana JC (2000) Skeletal muscle acidosis cor- relates with the severity of blood volume loss during shock and resuscitation. J Trauma Injury Infect Crit Care 49:1167–1173

21. Clavijo-Alvarez JA, Sims CA, Menconi M, Shim I, Ochoa C, Puyana JC (2004) Bladder mucosa pH and PCO

as a minimally invasive monitor of hemorrhagic shock and resusci- tation. J Trauma 57:1199–1210

22. Gutierrez G, Palizas F, Doglio G, et al (1992) Gastric intramucosal pH as a therapeutic

index of tissue oxygenation in critically ill patients. Lancet 339:195–199

23. Mohsenifar Z, Hay A, Hay J, Lewis MI, Koerner SK (1993) Gastric intramural pH as a pre- dictor of success or failure in weaning patients from mechanical ventilation. Ann Intern Med 119:794–798

24. Chance B, Cohen P, Jobsis F (1962) Intracellular oxidation-reduction state in vivo. Science 137:499–508

25. Van de Louw A, Cracco C, Cerf C, et al (2001) Accuracy of pulse oximetry in the intensive care unit. Intensive Care Med 27:1606–1613

26. The Acute Respiratory Distress Syndrome Network (2000) Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 342:1301–1308

27. Calzia E, Radermacher P (2003) Alveolar ventilation and pulmonary blood fl ow: the V

/Q concept. Intensive Care Med 29:1229–1232

28. Kumar A, Anel R, Bunnell E, et al (2004) Pulmonary artery occlusion pressure and central venous pressure fail to predict ventricular fi lling volume, cardiac performance, or the response to volume infusion in normal subjects. Crit Care Med 32:691–699

29. Thomas M, Shillingford J (1965) The circulatory response to a standard postural change in ischaemic heart disease. Br Heart J 27:17–27

30. Boulain T, Achard JM, Teboul JL, Richard C, Perrotin D, Ginies G (2002) Changes in blood pressure induced by passive leg raising predict response to fl uid loading in critically ill patients. Chest 121:1245–1252

31. Magder SA, Georgiadis G, Tuck C (1992) Respiratory variations in right atrial pressure predict response to fl uid challenge. J Crit Care 7:76–85

32. Kircher BJ, Himelman RB, Schiller NB (1990) Noninvasive estimation of right atrial pres- sure from the inspiratory collapse of the inferior vena cava. Am J Cardiol 66:493–496 33. Reuter DA, Bayerlein J, Goepfert MS, et al (2003) Infl uence of tidal volume on left ven-

tricular stroke volume variation measured by pulse contour analysis in mechanically ventilated patients. Intensive Care Med 29:476–480

34. Perel A (1997) Analogue values from invasive hemodynamic monitoring. In: Pinsky MR (ed) Applied Cardiovascular Physiology. Springer-Verlag, Berlin, pp 129–140

35. Perel A (1998) Assessing fl uid responsiveness by the systolic pressure variation in me- chanically ventilated patients. Systolic pressure variation as a guide to fl uid therapy in patients with sepsis-induced hypotension. Anesthesiology 89:1309–1310

36. Michard F, Boussat S, Chemla D, et al (2000) Relation between respiratory changes in arterial pulse pressure and fl uid responsiveness in septic patients with acute circulatory failure. Am J Respir Crit Care Med 162:134–138

37. Slama M, Masson H, Teboul JL, et al (2002) Respiratory variations of aortic VTI: a new index of hypovolemia and fl uid responsiveness. Am J Physiol Heart Circ Physiol 283:

H1729–H1733

38. Feissel M, Michard F, Mangin I, Ruyer O, Faller JP, Teboul JL (2001) Respiratory changes in aortic blood velocity as an indicator of fl uid responsiveness in ventilated patients with septic shock. Chest 119: 867–873

39. Pinsky MR (2004) Using ventilation-induced aortic pressure and fl ow variation to diag- nose preload responsiveness. Intensive Care Med 30:1008–1010

40. Creteur J, De Backer D, Vincent JL (1999) A dobutamine test can disclose hepatosplanch- nic hypoperfusion in septic patients. Am J Respir Crit Care Med 160:839–845

41. McKee M, Clarke A (1995) Guidelines, enthusiasms, uncertainty and the limits to pur- chasing. BMJ 310:101–104

42. Fenwick E, Wilson J, Sculpher M, Claxton K (2002) Pre-operative optimisation employing dopexamine or adrenaline for patients undergoing major elective surgery: a cost-effec- tiveness analysis. Intensive Care Med 28:599–608

43. Mythen MG, Webb AR (1995) Perioperative plasma volume expansion reduces the inci-

dence of gut mucosal hypoperfusion during cardiac surgery. Arch Surg 130:423–429

44. Venn R, Steele A, Richardson P, Poloniecki J, Grounds M, Newman P (2002) Randomised controlled trial to investigate infl uence of the fl uid challenge on duration of hospital stay and perioperative morbidity in patients with hip fractures. Br J Anaesth 88:65–71 45. McKendry M, McGloin H, Saberi D, Caudwell L, Brady AR, Singer M (2004) Randomised

controlled trial assessing the impact of a nurse delivered, fl ow monitored protocol for optimisation of circulatory status after cardiac surgery. BMJ 329:438–443

46. Yu M, Levy M, Smith P, Takiguchi S, Miyasaki A, Myers S (1993) Effect of maximizing oxygen delivery on morbidity and mortality rates in critically ill patients: a prospective randomized controlled trial. Crit Care Med 21:830–838

47. Pinsky MR (2002) Functional hemodynamic monitoring: Applied physiology at the bed-

side. In: Vincent JL (ed) Yearbook of Emergency and Intensive Care Medicine. Springer-

Verlag, Berlin pp 537–552